The study also examines factors that affect the sentences received by convicted defendants and uses economic logic to examine the critical factors underlying choices made and consequences generated. Study results indicate that factors increasing agreement between the prosecution and the accused also increase the likelihood of settlement. On the other hand, crime factors, such as the use of weapons or inflicting harm or death, increase the stakes and are more likely to reduce the probability of settlement. The study finds that while personal characteristics of the defendant usually do not affect the settlement decision, low levels of educational attainment increase the likelihood of trial. At trial, none of the variables relevant to settlement seems to have a significant effect on the determination of the defendant's guilt or innocence. The study also finds that plea bargaining leads to shorter sentences and that a defendant's personal characteristics influence type and length of sentences. Defendants with histories of drug abuse or mental illness receive longer sentences; females are often given shorter sentences. Race, education, and alcohol problems do not tend to influence punishment. The study illustrates the significance of economic models in predicting judicial outcomes. 20 footnotes.